Evaluation of selected recurrence measures in discriminating pre-ictal and inter-ictal periods from epileptic EEG data

Ngamga, E. J.; Bialonski, Stephan; Marwan, Norbert; Kurths, Jüergen; Geier, Christian; Lehnertz, Klaus

doi:10.1016/j.physleta.2016.02.024

Cited by 49 publications

(26 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These parameters usually increased in preictal state to the ictal state ( Fig 3 ). These changes, previously observed applying RQA analysis to EEG signals in epilepsy [ 58 , 59 ], suggest an increase in synchronization associated to seizures and could reflect the synchronization of neurons during seizures. Consistently CosEn is one of the most frequently selected features.…”

Section: Discussionmentioning

confidence: 52%

Patient-specific seizure prediction based on heart rate variability and recurrence quantification analysis

et al. 2018

View full text Add to dashboard Cite

Epilepsy is often associated with modifications in autonomic nervous system, which usually precede the onset of seizures of several minutes. Thus, there is a great interest in identifying these modifications enough time in advance to prevent a dangerous effect and to intervene. In addition, these changes can be a risk factor for epileptic patients and can increase the possibility of death. Notably autonomic changes associated to seizures are highly depended of seizure type, localization and lateralization. The aim of this study was to develop a patient-specific approach to predict seizures using electrocardiogram (ECG) features. Specifically, from the RR series, both time and frequency variables and features obtained by the recurrence quantification analysis were used. The algorithm was applied in a dataset of 15 patients with 38 different types of seizures. A feature selection step, was used to identify those features that were more significant in discriminating preictal and interictal phases. A preictal interval of 15 minutes was selected. A support vector machine (SVM) classifier was then built to classify preictal and interictal phases. First, a classifier was set up to classify preictal and interictal segments of each patient and an average sensibility of 89.06% was obtained, with a number of false positive per hour (FP/h) of 0.41. Then, in those patients who had at least 3 seizures, a double-cross-validation approach was used to predict unseen seizures on the basis of a training on previous ones. The results were quite variable according to seizure type, achieving the best performance in patients with more stereotypical seizure. The results of the proposed approach show that it is feasible to predict seizure in advance, considering patient-specific, and possible seizure specific, characteristics.

show abstract

Section: Discussionmentioning

confidence: 52%

Patient-specific seizure prediction based on heart rate variability and recurrence quantification analysis

et al. 2018

View full text Add to dashboard Cite

show abstract

“…The Mean Recurrence Time (RT) [ 78 – 79 ] is an alternative estimator for T2 but for the calculation, the focus is put on the white vertical lines [ 45 ]. Vertical and horizontal white bands result from rarely occurring states [ 80 ].…”

Section: Methodsmentioning

confidence: 99%

“…First they calculate the local clustering coefficient for each node, and the clustering coefficient is the average of all nodes [ 81 ]. The adjacency matrix A used in their definitions is the recurrence matrix from which the identity matrix is subtracted ( A ij = R ij − δ ij where δ ij is the Kronecker delta) [ 79 ].…”

Section: Methodsmentioning

confidence: 99%

Improving the understanding of sleep apnea characterization using Recurrence Quantification Analysis by defining overall acceptable values for the dimensionality of the system, the delay, and the distance threshold

et al. 2018

View full text Add to dashboard Cite

Our contribution focuses on the characterization of sleep apnea from a cardiac rate point of view, using Recurrence Quantification Analysis (RQA), based on a Heart Rate Variability (HRV) feature selection process. Three parameters are crucial in RQA: those related to the embedding process (dimension and delay) and the threshold distance. There are no overall accepted parameters for the study of HRV using RQA in sleep apnea. We focus on finding an overall acceptable combination, sweeping a range of values for each of them simultaneously. Together with the commonly used RQA measures, we include features related to recurrence times, and features originating in the complex network theory. To the best of our knowledge, no author has used them all for sleep apnea previously. The best performing feature subset is entered into a Linear Discriminant classifier. The best results in the “Apnea-ECG Physionet database” and the “HuGCDN2014 database” are, according to the area under the receiver operating characteristic curve, 0.93 (Accuracy: 86.33%) and 0.86 (Accuracy: 84.18%), respectively. Our system outperforms, using a relatively small set of features, previously existing studies in the context of sleep apnea. We conclude that working with dimensions around 7–8 and delays about 4–5, and using for the threshold distance the Fixed Amount of Nearest Neighbours (FAN) method with 5% of neighbours, yield the best results. Therefore, we would recommend these reference values for future work when applying RQA to the analysis of HRV in sleep apnea. We also conclude that, together with the commonly used vertical and diagonal RQA measures, there are newly used features that contribute valuable information for apnea minutes discrimination. Therefore, they are especially interesting for characterization purposes. Using two different databases supports that the conclusions reached are potentially generalizable, and are not limited by database variability.

show abstract

“…Previous studies have examined several RQA values (RR, DET, LAM) computed on a single scale from EEG time series. Determinism (DET) and Laminarity (LAM) were been found to be lower closer to the epileptic zone in patients with focal epilepsies, while recurrence rate (RR) and entropy were higher [56]. This suggests that similar differences in general might be found between AE cases and controls, since AE is not localized.…”

Section: Resultsmentioning

confidence: 97%

Nonlinear EEG biomarker profiles for autism and absence epilepsy

Bosl

Loddenkemper

Nelson

2017

Neuropsychiatr Electrophysiol

View full text Add to dashboard Cite

Background: Although autism and epilepsy are considered to be different disorders, epileptiform EEG activity is common in people with autism even when overt seizures are not present. The relatively high comorbidity between autism and all epilepsy syndromes suggests the possibility of common underlying neurophysiological mechanisms. Although many different epilepsies may be comorbid with autism, absence epilepsy is a generalized epilepsy syndrome with seizures that appear as staring spells, with no motor signs and no focal lesions, making it more difficult to diagnose. Application of nonlinear methods for EEG signal analysis may enable characterization of brain activity that can help to delineate neurophysiological commonalities and differences between autism and epilepsy. Multiscale entropy and recurrence quantitative analysis (RQA) were computed from EEG signals derived from children with autism or absence epilepsy and compared with the goal of finding significant and potentially clinically useful biomarkers neurophysiological differences between these two childhood disorders. Methods: Multiscale entropy and a multiscale version of RQA were computed from EEG data obtained from 92 children were collected in two different settings at Boston Children's Hospital. Short segments of alert resting state EEG were selected for analysis. A complexity index derived from entropy and RQA methods was computed from each of 19 standard EEG channels for all subjects using publicly available software. Statistical comparisons were made between the groups. Machine learning classifiers were also used to determine which derived features were most significantly different among the groups, and to determine classification specificity and sensitivity. Results: Significant differences were found between absence, autism, and control groups in a number of different scalp locations and the values of complexity index. Autism values appeared to be intermediate between epilepsy and control in many locations, and differences between controls and absence patients were more widely distributed across scalp locations. Classification algorithms were able to distinguish absence epilepsy and autism cases from controls with high (>95%) accuracy. Importantly, two independent control groups, although they were derived from different settings and with different equipment were statistically indistinguishable.

show abstract

Evaluation of selected recurrence measures in discriminating pre-ictal and inter-ictal periods from epileptic EEG data

Cited by 49 publications

References 67 publications

Patient-specific seizure prediction based on heart rate variability and recurrence quantification analysis

Patient-specific seizure prediction based on heart rate variability and recurrence quantification analysis

Improving the understanding of sleep apnea characterization using Recurrence Quantification Analysis by defining overall acceptable values for the dimensionality of the system, the delay, and the distance threshold

Nonlinear EEG biomarker profiles for autism and absence epilepsy

Contact Info

Product

Resources

About